from __future__ import annotations import json import time import urllib.error import urllib.request from typing import Any, Callable from datahub.adapters.base import AdapterError, MarketAdapter from datahub.normalize import ( normalize_auction, normalize_calendar, normalize_daily, normalize_index_daily, normalize_moneyflow, normalize_stock, normalize_valuation, ) TUSHARE_URL = "http://api.tushare.pro" TUSHARE_FIELDS = { "trade_cal": "exchange,cal_date,is_open,pretrade_date", "stock_basic": "ts_code,symbol,name,area,industry,market,list_status,list_date", "daily": "ts_code,trade_date,open,high,low,close,pct_chg,vol,amount", "daily_basic": "ts_code,trade_date,turnover_rate,volume_ratio,total_mv,circ_mv,pe_ttm,pb,ps_ttm,dv_ttm", "adj_factor": "ts_code,trade_date,adj_factor", "index_daily": "ts_code,trade_date,open,high,low,close,pct_chg,vol,amount", "moneyflow": ( "ts_code,trade_date,buy_sm_amount,sell_sm_amount,buy_md_amount,sell_md_amount," "buy_lg_amount,sell_lg_amount,buy_elg_amount,sell_elg_amount,net_mf_amount" ), "stk_auction": "ts_code,trade_date,vol,price,amount,pre_close,turnover_rate,volume_ratio,float_share", } DATASET_API = { "calendar": "trade_cal", "stocks": "stock_basic", "daily": "daily", "valuation": "daily_basic", "adj_factor": "adj_factor", "index_daily": "index_daily", "moneyflow": "moneyflow", "auction": "stk_auction", } # Website actual index usage: market cards / 90-day charts (SH/SZ/CYB) plus # screener 沪深300 benchmark (lookback up to 260 trading days). WEBSITE_INDEX_CODES = ("000001.SH", "399001.SZ", "399006.SZ", "000300.SH") DEFAULT_INDEX_CODES = WEBSITE_INDEX_CODES class TushareAdapter(MarketAdapter): name = "tushare" def __init__( self, token: str, timeout: int = 30, transport: Callable[[str, dict[str, Any], str], list[dict[str, Any]]] | None = None, ) -> None: self.token = token self.timeout = timeout self._transport = transport def probe(self) -> dict[str, Any]: if not self.token: return {"provider": self.name, "configured": False, "state": "unconfigured"} started = time.perf_counter() try: rows = self.fetch("calendar", {"exchange": "SSE", "start_date": "20200102", "end_date": "20200102"}) except AdapterError as exc: return { "provider": self.name, "configured": True, "state": "error", "message": str(exc), "latency_ms": round((time.perf_counter() - started) * 1000), } return { "provider": self.name, "configured": True, "state": "ok" if rows else "empty", "latency_ms": round((time.perf_counter() - started) * 1000), } def fetch(self, dataset: str, params: dict[str, Any]) -> list[dict[str, Any]]: api_name = DATASET_API.get(dataset, dataset) fields = TUSHARE_FIELDS.get(api_name, "") query_params = dict(params) if api_name == "stock_basic" and "list_status" not in query_params: query_params["list_status"] = "L" if api_name == "trade_cal" and "exchange" not in query_params: query_params["exchange"] = "SSE" if api_name == "index_daily" and "ts_code" not in query_params: # Caller typically loops codes; a missing code would pull nothing useful. query_params.setdefault("ts_code", DEFAULT_INDEX_CODES[0]) return self._query(api_name, query_params, fields) def fetch_index_daily(self, trade_date: str, codes: tuple[str, ...] = DEFAULT_INDEX_CODES) -> list[dict[str, Any]]: rows: list[dict[str, Any]] = [] for ts_code in codes: rows.extend(self.fetch("index_daily", {"ts_code": ts_code, "trade_date": trade_date})) return rows def normalize(self, dataset: str, rows: list[dict[str, Any]]) -> list[dict[str, Any]]: mapping = { "calendar": normalize_calendar, "trade_cal": normalize_calendar, "stocks": normalize_stock, "stock_basic": normalize_stock, "daily": normalize_daily, "valuation": normalize_valuation, "daily_basic": normalize_valuation, "moneyflow": normalize_moneyflow, "auction": normalize_auction, "stk_auction": normalize_auction, "index_daily": normalize_index_daily, } fn = mapping.get(dataset) if fn is None: if dataset == "adj_factor": return [ { "ts_code": str(row.get("ts_code") or "").upper(), "trade_date": str(row.get("trade_date") or ""), "adj_factor": row.get("adj_factor"), } for row in rows ] raise AdapterError(f"unsupported dataset: {dataset}") return [fn(row) for row in rows] def _query(self, api_name: str, params: dict[str, Any], fields: str) -> list[dict[str, Any]]: if self._transport is not None: return self._transport(api_name, params, fields) if not self.token: raise AdapterError("Tushare token 未配置") payload = json.dumps( {"api_name": api_name, "token": self.token, "params": params, "fields": fields} ).encode("utf-8") request = urllib.request.Request( TUSHARE_URL, data=payload, headers={"Content-Type": "application/json", "User-Agent": "XiaobaiDatahub/0.1"}, method="POST", ) try: with urllib.request.urlopen(request, timeout=self.timeout) as response: result = json.loads(response.read().decode("utf-8")) except json.JSONDecodeError: raise AdapterError("Tushare returned invalid json") from None except (urllib.error.URLError, TimeoutError) as exc: raise AdapterError(f"Tushare request failed: {exc}") from exc if result.get("code") != 0: raise AdapterError(result.get("msg") or "Tushare returned an unknown error") data = result.get("data") or {} columns = data.get("fields") or [] return [dict(zip(columns, item)) for item in data.get("items") or []]